Evaggelos G. Kardakos

Application of time series and artificial neural network models in short-term forecasting of PV power generation

This paper addresses two practical methods for electricity generation forecasting of grid-connected PV plants. The first model is based on seasonal ARIMA time-series analysis and is further improved by incorporating short-term solar radiation forecasts derived from NWP models. The second model adopts artificial neural networks with multiple inputs. Day-ahead and rolling intra-day forecast updates are implemented to evaluate the forecasting errors. All models are compared in terms of the Normalized (with respect to the PV installed capacity) Root Mean Square Error (NRMSE). Simulation results from the application of the forecasting models in different PV plants of the Greek power system are presented.

A unified unit commitment — Economic dispatch model for short-term power system scheduling under high wind energy penetration

This paper presents a unified unit commitment and economic dispatch tool for the short-term scheduling of a power system under high renewable penetration. The proposed model uses variable time resolution and scheduling horizon extended up to 36 hours ahead and produces robust real-time decisions making the short-term operation of the power system almost insensitive to RES forecast errors. The proposed model is tested for a monthly period on the Greek interconnected power system using real load and wind power data for two different wind penetration levels. Simulation results show that the proposed methodology allows for the accommodation of large amounts of wind energy into the short-term scheduling of the power system at minimum cost.

Comparison of SARIMAX, SARIMA, modified SARIMA and ANN-based models for short-term PV generation forecasting

This paper compares four practical methods for electricity generation forecasting of grid-connected Photovoltaic (PV) plants, namely Seasonal Autoregressive Integrated Moving Average (SARIMA) modeling, SARIMAX modeling (SARIMA modeling with exogenous factor), modified SARIMA modeling, as a result of an a posteriori modification of the SARIMA model, and ANN-based modeling. Interesting results regarding the necessity and the advantages of using exogenous factors in a time series model are concluded from this comparison. Finally, intra-day forecasts updates are implemented to evaluate the forecasting errors of the SARIMA and the SARIMAX models. Their comparison highlights differences in accuracy between the two models. All models are compared in terms of the Normalized (with respect to the PV installed capacity) Root Mean Square Error (NRMSE) criterion. Simulation results from the application of the forecasting models in a PV plant in Greece using real-world data are presented.

Large-scale res integration in electricity markets: Challenges and potential solutions

The increasing shares of renewable energy in power systems have a significant impact on the operation of electricity markets and grids worldwide. This paper provides an overview of the main challenges that high shares of renewable generation introduce in the power system management and electricity markets operation as well as a brief description of potential solutions for alleviating the negative implications caused by the large-scale renewable integration. Additionally, a summary of the core research activity that has been performed in the context of a relevant academic research project, called “Large-Scale Renewable Integration in Electricity Markets”, (acronym, “LaRInEM”) is presented and valuable conclusions regarding the efficient integration of large amounts of renewable energy in electricity markets are drawn.

Benefits of demand response on a wind power producer bidding strategy

This paper proposes an optimal bidding strategy for a wind producer who participates as price-maker in the day-ahead electricity market. In order to empower the wind producer against the challenges introduced by the intermittent nature of wind power and improve his position into the market, demand response schemes are considered to be incorporated into his portfolio. The producer objective is the maximization of the day-ahead profit in conjunction with the minimization of the anticipated real-time production imbalance charges. A two-stage stochastic bi-level optimization model has been formulated, where the uncertainty lies in the day-ahead wind production and load consumption. Simulations on the Greek power system prove the effectiveness of the proposed synergy.

Artificial neural network-based methodology for short-term electric load scenario generation

In this paper a novel scenario generation methodology based on artificial neural networks (ANNs) is proposed. The methodology is able to create scenarios for various power system-related stochastic variables. Scenario reduction methodologies can then be applied to effectively reduce the number of scenarios. An application of the methodology for the creation of short-term electric load scenarios for one day up to seven days ahead is presented. Test results on the real-world insular power system of Crete present the effectiveness of the proposed methodology.